similar to: [LLVMdev] [RFC] AArch64: Should we disable GlobalMerge?

Summary ======= I'm planning on adjusting SimplifyCFG so that it doesn't turn two-entry phi nodes into selects until later in the pass pipeline, to give passes which can understand phis but not selects more opportunity to optimize. The thing I'm trying to do which made me think of doing this is described below, but from the benchmarking I've done it looks like this is overall a

I'm concerned that we're focusing on one side of this.  Let me point out a few concerns w/changing the canonical form here: 1. LICM does not know how to hoist or sink regions.  It does know how to hoist and sink selects. 2. InstCombine has limited support for triangles/diamonds, but fairly extensive support for selects. 3. EarlyCSE and GVN do not know how to eliminate fully

> On Aug 15, 2018, at 10:57 PM, Hal Finkel via llvm-dev <llvm-dev at lists.llvm.org> wrote: > > > On 08/15/2018 02:38 PM, Philip Reames via llvm-dev wrote: >> I'm concerned that we're focusing on one side of this. Let me point out a few concerns w/changing the canonical form here: >> >> LICM does not know how to hoist or sink regions. It does know

Hi, LLVM’s SLP-vectorizer is a new pass that combines similar independent instructions in a straight-line code. It is currently not enabled by default, and people who want to experiment with it can use the clang command line flag “-fslp-vectorize”. I ran LLVM’s test suite with and without the SLP vectorizer on a Sandybridge mac (using SSE4, w/o AVX). Based on my performance measurements

Hi, Below you can see the updated benchmark results for the new SLP-vectorizer. As you can see, there is a small number of compile time regressions, a single major runtime *regression, and many performance gains. There is a tiny increase in code size: 30k for the whole test-suite. Based on the numbers below I would like to enable the SLP-vectorizer by default for -O3. Please let me know if you

On Jul 14, 2013, at 9:52 PM, Chris Lattner <clattner at apple.com> wrote: > > On Jul 13, 2013, at 11:30 PM, Nadav Rotem <nrotem at apple.com> wrote: > >> Hi, >> >> LLVM’s SLP-vectorizer is a new pass that combines similar independent instructions in a straight-line code. It is currently not enabled by default, and people who want to experiment with it

On Jul 13, 2013, at 11:30 PM, Nadav Rotem <nrotem at apple.com> wrote: > Hi, > > LLVM’s SLP-vectorizer is a new pass that combines similar independent instructions in a straight-line code. It is currently not enabled by default, and people who want to experiment with it can use the clang command line flag “-fslp-vectorize”. I ran LLVM’s test suite with and without the SLP

tl;dr in low data situations we don’t look at past information, and that increases the false positive regression rate. We should look at the possibly incorrect recent past runs to fix that. Motivation: LNT’s current regression detection system has false positive rate that is too high to make it useful. With test suites as large as the llvm “test-suite” a single report will show hundreds of

Hi, Sorry for the delay in response. I measured the code size change and noticed small changes in both directions for individual programs. I found a 30k binary size growth for the entire testsuite + SPEC. I attached an updated performance report that includes both compile time and performance measurements. Thanks, Nadav On Jul 14, 2013, at 10:55 PM, Nadav Rotem <nrotem at apple.com>

Hello all, I have evaluated the compile-time and execution-time performance of Polly canonicalization passes. Details can be referred to http://188.40.87.11:8000/db_default/v4/nts/recent_activity. There are four runs: pollyBasic (run 45): clang -O3 -Xclang -load -Xclang LLVMPolly.so pollyNoGenSCEV (run 44): clang -O3 -Xclang -load -Xclang LLVMPolly.so -mllvm -polly -mllvm -polly-codegen-scev

At 2013-09-09 13:07:07,"Tobias Grosser" <tobias at grosser.es> wrote: >On 09/09/2013 05:18 AM, Star Tan wrote: >> >> At 2013-09-09 05:52:35,"Tobias Grosser" <tobias at grosser.es> wrote: >> >>> On 09/08/2013 08:03 PM, Star Tan wrote: >>> Also, I wonder if your runs include the dependence analysis. If this is >>> the

Hello all, I have done some basic experiments about Polly canonicalization passes and I found the SCEV canonicalization has significant impact on both compile-time and execution-time performance. Detailed results for SCEV and default canonicalization can be viewed on: http://188.40.87.11:8000/db_default/v4/nts/32 (or 33, 34) *pNoGen with SCEV canonicalization (run 32): -O3 -Xclang -load

Are these 225 compile time regressions real? It sure looks bad! Ciao, Duncan. On 01/12/11 09:39, llvm-testresults at cs.uiuc.edu wrote: > > bwilson__llvm-gcc_PROD__i386 nightly tester results > > URL http://llvm.org/perf/db_default/simple/nts/380/ > Nickname bwilson__llvm-gcc_PROD__i386:4 > Name curlew.apple.com > > Run ID Order Start Time End Time > Current 380

Hi, I have just tagged 3.7.1-rc1, so it is ready for testing. As a reminder, when doing regression testing, use the 3.7.0 release as your baseline. Thanks, Tom

Hi Chandler, Thanks for fixing the problem with the insertps mask. Generally the new shuffle lowering looks promising, however there are some cases where the codegen is now worse causing runtime performance regressions in some of our internal codebase. You have already mentioned how the new shuffle lowering is missing some features; for example, you explicitly said that we currently lack of

On Sun, Jan 20, 2013 at 1:26 PM, Redmond, Paul <paul.redmond at intel.com> wrote: > There is almost certainly a bug in lnt or the makefiles. > > I changed the body of Burg main to the following: > > + printf("Hello World\n"); > + return 0; > > > I re-ran the test-suite again and got the following errors: > > --- Tested: 986 tests -- > FAIL:

There is almost certainly a bug in lnt or the makefiles. I changed the body of Burg main to the following: + printf("Hello World\n"); + return 0; I re-ran the test-suite again and got the following errors: --- Tested: 986 tests -- FAIL: MultiSource/Applications/Burg/burg.execution_time (494 of 986) FAIL: MultiSource/Applications/ClamAV/clamscan.execution_time (495 of 986) FAIL:

Hi Bob, are these performance regressions real? They look pretty serious. Ciao, Duncan. On 10/12/11 09:40, llvm-testresults at cs.uiuc.edu wrote: > > bwilson__llvm-gcc_PROD__i386 nightly tester results > > URL http://llvm.org/perf/db_default/simple/nts/332/ > Nickname bwilson__llvm-gcc_PROD__i386:4 > Name curlew.apple.com > > Run ID Order Start Time End Time >

A big compile time regression. Any ideas? Ciao, Duncan. On 22/07/11 19:13, llvm-testresults at cs.uiuc.edu wrote: > > bwilson__llvm-gcc_PROD__i386 nightly tester results > > URL http://llvm.org/perf/db_default/simple/nts/253/ > Nickname bwilson__llvm-gcc_PROD__i386:4 > Name curlew.apple.com > > Run ID Order Start Time End Time > Current 253 0 2011-07-22 16:22:04

Yes, they are real. I re-ran the two tests with the biggest execution time regressions, and the results were completely reproducible. On Oct 12, 2011, at 1:24 AM, Duncan Sands wrote: > Hi Bob, are these performance regressions real? They look pretty serious. > > Ciao, Duncan. > > On 10/12/11 09:40, llvm-testresults at cs.uiuc.edu wrote: >> >>